MaxTheSurvivor

Hi devs, I'd like to move my test server 1.0.150192 (latest version at this time) into a pre-production environment now that my huge base building tests are finished and unlike the previous versions it hasn't crashed yet. I have a second steam account and I'd like to be able to run the DayZ Server without buying a second time the game since I'd like to install it on a VM on one of my esxi hosts. The tests ran fine this time, they started on the 5th of January to this day and each time I made some major modifications to the base I made a backup of the "dayzOffline.chernarusplus" folder (backups available for you if you need them for analyzing how persistence behaves each time we add more stuff in). The test server is non-modded and I'd like to open the server to other players (only a few will be allowed in these final persistence test stages) so that they can mess around with persistence themselves and report to me the last actions they've done in case of a catastrophic crash. The problem is that for the moment the servers can only be running on the same computer as the DayZ Client, in other words my gaming rig. This is not what I would call a controlled test environment since gaming computers usually have a lot of processes like LED software control and other bloatware that could cause crashes of the DayZ server without the latter being the culprit. This is too complex of an environment to run viable tests. And tbh I don't think people renting servers will have the patience to run tests that spread over several weeks and hundreds of hours of gaming potentially hunting nothing but rumors. This kind of time-consuming tests should be offloaded to community members that want to give a hand by presenting you with precise data to analyze (like regular backups). I might be wrong but I haven't even seen a single thread talking of persistence, how to test it and how to run the server in a controlled environment (launch options, logs, CPU model, RAM, etc..) so that you guys have real life/production server pre-processed data to analyze instead of having you running multiple servers for hundreds of hours in hope that something interesting persistence related happens.

Yep that's this sentence that put the doubt in my mind, I understood it as "You have to have DayZ Client on the same computer in order to run the server", basically forcing people to run the server on their gaming computer. But I was wrong and now I can leave the server running 24/7 on my virtualization servers. I wasn't sure we were allowed to run the server "outside" of steam and feared a BattleEye ban but the above comments clarified this for me.

The server is up and running, thanks for your help. I wasn't sure I could simply move the files to the server without having the client installed. Time to add the NAT and firewall rules and it'll be good to go. Thumbs up guys

Hi chili1496, There's no real advantage to building a base. As eno stated bases are easy to break-in on vanilla servers. There's a mod to make bases pretty much indestructible but the community server has to run the mod. Best case scenario; bases are indestructible and persistence is reliable at 100% = more work for you. Bases are just more work, something to do beside the usual run & gun routine. Base building is very time and resources consuming and requires you already have a few tents filled with base building materials and tools ready for when you found the perfect spot to build your base. You'd better have 20 rolls of duct tape and 50 boxes of nails ready or base building will be very very very long. TBH the most difficult in base building is to find the right spot for your needs, in other words what do you want a base for? Will it be a place where you can come grab gear in case you win a free ticket back to the coast? In this case the base should be tiny/discreet and relatively close to the coast. Will it be a base to "control" the area? In this case you'll need more room for a bigger base and you'll need people to give you a hand and keep it running and protected from other players. Choosing the right spot to build a base is not that easy. Terrain that looks flat for you will most of the time not be flat enough for the game and chances are that you won't be able to close the perimeter forcing you to abandon the construction or adapt the shape of your base (a real nightmare). A good trick to know if a terrain is flat enough is to find a car tent and try to place it where you want to build your base. If you can pitch your tent anywhere it means the terrain is flat enough and you can try to build something there. If even before raising the first wall you find out that you cannot even pitch a tent despite being in the middle of a huge field it means the terrain is not flat enough (the slightest of slopes is enough to make a visually perfect flat terrain unfit for construction).

Thanks for all your replies guys. OK, I haven't seen any documentation telling me that I could run the server without the DayZ Client. I'll try that today see if I can make it work. Beans for you all

Home sweet home

You're wrong about this, the problem with persistence is because of the enormous number of items in game. You probably noticed that modded servers that add the old guns and other clothing are much more unstable than the vanilla ones. Adding "content" aka guns will only make servers even more unstable. Persistence is not only related to base building but to all in-game objects. Back in 2014 the only persistent items were those you were carrying in your inventory, all other items were deleted and replaced upon server restart. My biggest fear at that time was a server restart just when I was about to pick up the elusive M4 because I knew its existence wasn't persistent. As long as DayZ Server stores items in a file instead of a proper DB we'll have growing instability as more items are generated and stored. And for security/anti-cheat reasons I don't see BI changing their mind and adding a DB server to store game data, it would be too easy to alter the DB contents possibly even while the server is running to delete or add items in-game like removing the gun you have in hands. Persistence is THE ONLY REAL priority among all the top priorities because all the others are more or less dependent on how the server is able to handle objects without crashing.

It's not your imagination. Loot indeed cycles, from what I observed loot that has a better value will spawn for a specific amount of days before not spawning anymore; high capacity vests, cargo pants, m65 field jacket (especially the black ones) will spawn normally and all of the sudden you can't find them anymore. I have no specific data to backup my claims but observation tells me there's definitely a loot cycle. Is it the <cost> in types.xml (I don't think so since all seem to be at 100) or the <restock>, some are set at 0 which makes me think that once <lifetime> has reached its maximum it will check the <restock> value that is probably in minutes and will wait unit the <restock> has passed to re-spawn an item. If the <restock> value is set to 0 it will immediately respawn an item if the value of the total amount of this kind of item is lower than the <min> value for the whole server. <max> value is probably only used at the first start of the server or at the start of the server if persistance is disabled in the server config file. Note that this is just my personal theory as I have no evidence to back up my claims ("Always have a disclaimer before some fool considers you accountable of his own mistakes" by myself, now, here)

I agree, combat logging is something that ruins the game and I don't want some complicated solution when all you need is to increase the disconnect timer. 60 seconds seems fine, enough to scare server hoppers and prevent players to leave the battlefield in the middle of a fight.

It's persistence all together that is not working. Base building only accelerates the process since you create new items. I ran long tests on my test server (around 100 hours) and had 10 watchtowers, 15 tents, barrels, as I was building the base (half-way before crash) and adding more and more items in tents and barrels. With the planks and metal sheets dup bug I generated even more items that all had to be persistant. Slowly but surely the server became more and more unstable, having brutal fps drops and lags until the final crash wiped all persistant items on the map, all of them, not just those around me. --> Also know that tents and barrels probably need to be "reactivated" periodically like we had to 3 years ago. It was something like: 8 days before an untouched barrel even with items inside would be wiped and you had to pay a visit to all your tents and barrels and move an item out of it then put it back to reactivate the timer. I think people are sometimes mistaken and consider features as bugs. The fact that barrels and tents "despawn" after a specific amount of time is not a bug and it's by design. As always, I suggest you run tests on your server; tell your people to not touch a barrel or a tent, just leave one of them alone with one item inside to see how long it will take for it to vanish. You can check this page https://dayz.gamepedia.com/Loot that lists the storage items and the time it takes before they despawn if untouched. I don't know if it's still the case though.

Dude I built a huge base, I mean massive, about 100 hundred hours of work, half-way through (10 watchtowers, 15 tents, fences between the towers) and the persistance crash wiped everything. 1.0 was supposed to be a stable release with the core mechanics being reliable, persistance is not, cars are not, what remains of that 1.0? Not much. Well, I supported the game until the 1.0 release, now I'll start to be a little less kind with the devs. They wanted a rushed 1.0? They can't hide anymore behind the "it's alpha", there's no more "I understand" buton.

Do not do base building! I just finished a 100 hours test on my private test server. Obviously I was alone on it. I tested base building by building a huge base: 15 tents, 10 watchtowers with fences in-between, I was half-way through of finishing the base, the more base parts I was building the more lags I had, at some point the server started to crash every 3-4 hours and today I had a crash that resulted in the entire base being wiped. I think servers can't handle more that a few (around 30-40) base building items before persistance goes south and ends up wiping everything you've built. Server backups can save the day but since the crashes frequency increases with the amount of persistant items on the server it means you'll have to take down the server every 2 hours to do a full backup. That's something you can do you when you're alone on a test server but not if you want to have 40-60 players having an enjoyable experience. The server was running on a i7-8700K @4.7 on all cores, 16GB of DDR4 3200 RAM and an nvme SSD drive, so plenty of resources for the server to run smooth.

Hi, the last patch ruined once more performance. I had 120 fps in average with my 1080ti @ 1440p ultra-wide, now it's around 90 fps. I was playing the game 3 hours ago, went back home, updated the server files and client and performance has dropped significantly, just like after the infamous 22nd of Nov patch. anyone else noticed some heavy fps drop?

I'm validating all the files once more, just in case.

I have a 1080ti and playing at 1440p ultra-wide (3440x1440) with a i7-8700K @4.7 on all cores and nvme ssd. 3 hours before the patch I had an average of 120fps when I logged off, after the patch I'm getting around 90 fps with the same settings.

Yeah, I just lost 30 fps again, just like after the 22nd of Nov infamous patch. Glad I'm not the only one.

Don't bother, the devs are getting roasted on twitter, it's over, they failed the launch, now the servers will have 0.2 players in average

server is crashing, client is crashing right after "respawning in 15 seconds"

they're pushing the client update now

https://www.dayzspy.com/

Already done that 40 mins ago, 30 mins ago, 20 mins ago, 10 mins ago, and wait 60 seconds ago

they're pushing a new patch for the server

Houston, we have a problem

Great, how are we supposed to play the game then?